Epic, Cerner electronic health records get free text transcription from smartphones and tablets; computerized physician order entry and natural language processing are in the works.
IW 500: 10 Healthcare IT Innovators
(click image for larger view and for slideshow)
The new mobile-native electronic health records (EHRs) systems of Epic and Cerner are being voice-enabled via recent deals with Nuance Communications. Allscripts also incorporates speech recognition from Nuance, as well as MModal, in its EHR applications for mobile devices.
When Nuance's cloud-based speech recognition program is imbedded in Epic or Cerner EHRs, physicians who use them will be able to dictate into their iPad or iPhone and have the machine-transcribed text pop up in the correct section of their EHR. This is something that physicians have long been able to do with conventional EHRs that they use on desktop or laptop computers. But many doctors want to do it on their mobile devices to facilitate documentation.
According to a Nuance press release, Epic's Haiku application for the iPhone and Canto software for the iPad will connect to Nuance's cloud-based server via an interface with users' desktop EHRs. "This connectivity will provide maintenance-free access and automatic updates of personal speech profiles, medical vocabularies, and user-defined shortcuts. Customers can also enter progress notes, histories, and physical exams via voice," the announcement said.
Because Nuance's platform is cloud-based, "it's form-factor neutral" and can be used with iOS, Android, or other mobile or desktop devices, as well as thick- and thin-client systems, said Joe Petro, senior VP of R&D for Nuance's healthcare division, in an interview with InformationWeek Healthcare.
While Epic and Cerner are deploying Nuance only for speech recognition, he said, users can enable the "command and control" aspects of Nuance's Dragon program to fill out forms in an EHR. "You can navigate the form in a natural way and ratchet through the fields while narrating."
Nevertheless, he noted, this still produces free text notes without entering data in discrete fields. Nuance offers "clinical language understanding" (CLU) capability (otherwise known as natural language processing), but Petro admitted that it's not quite ready to extract data from speech and place it in specific fields reliably.
Physicians have long been able to prescribe electronically on mobile devices, but it requires them to enter orders manually. Petro said that Nuance's speech recognition is accurate enough to recognize simple medication orders. "With Epic, for example, you could order '250 mg of amoxicillin by mouth three times a day.' And if you deploy the right kind of interface, as you're speaking that, we're teasing the order apart with CLU and placing it in the right places."
However, he added that speech recognition on mobile devices is not equal to the more complex tasks involved in computerized physician order entry (CPOE). "There's nothing technically to prevent you from using it for CPOE," he said. "But CPOE can get very sophisticated, and generally the form factor on the iPhone or iPad gets in the way of the most sophisticated orders."
The other problem with using voice recognition programs in ordering is that the application might misinterpret the name or dosage of a particular medication, which could be dangerous in patient care. But Petro said that Nuance is looking to use an "intermediate interface" that allows a doctor to check on whether the computer has accurately translated his spoken words before he pushes the button to place an order or record the text in his EHR.
Allscripts, Cerner, and Epic--the major vendors that are now offering mobile-native EHRs--are all planning to allow clinicians to order simple prescriptions or lab tests by voice in those applications, Petro noted. But they have a few other priorities to take care of first.
"The first step is to make these devices as productive as they can be," he said. "And because of the [mobile device] form factor and the virtual keyboard, text entry is the first thing they have to crack. That said, all these companies are six-to-18 months out in terms of developing [ordering by voice]. They're all contemplating the ability to use voice to drive these things."
InformationWeek Healthcare brought together eight top IT execs to discuss BYOD, Meaningful Use, accountable care, and other contentious issues. Also in the new, all-digital CIO Roundtable issue: Why use IT systems to help cut medical costs if physicians ignore the cost of the care they provide? (Free with registration.)
InformationWeek Tech Digest, Nov. 10, 2014Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?